Factors that impact the loan approval for an applicant

Investigation Overview:

This project is about the factors or features that could help the loan applicant to get the approval status.

Dataset Overview:

This data set contains 113,937 loans with 81 variables on each loan, including loan amount, borrower rate (or interest rate), current loan status, borrower income, and many others.

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
%matplotlib inline
from functools import reduce
from datetime import datetime
import seaborn as sns
import plotly.express as px
import plotly.graph_objects as go

df = pd.read_csv('prosperLoanData.csv')
display(df.head())

	ListingKey	ListingNumber	ListingCreationDate	CreditGrade	Term	LoanStatus	ClosedDate	BorrowerAPR	BorrowerRate	LenderYield	...	LP_ServiceFees	LP_CollectionFees	LP_GrossPrincipalLoss	LP_NetPrincipalLoss	LP_NonPrincipalRecoverypayments	PercentFunded	Recommendations	InvestmentFromFriendsCount	InvestmentFromFriendsAmount	Investors
0	1021339766868145413AB3B	193129	2007-08-26 19:09:29.263000000	C	36	Completed	2009-08-14 00:00:00	0.16516	0.1580	0.1380	...	-133.18	0.0	0.0	0.0	0.0	1.0	0	0	0.0	258
1	10273602499503308B223C1	1209647	2014-02-27 08:28:07.900000000	NaN	36	Current	NaN	0.12016	0.0920	0.0820	...	0.00	0.0	0.0	0.0	0.0	1.0	0	0	0.0	1
2	0EE9337825851032864889A	81716	2007-01-05 15:00:47.090000000	HR	36	Completed	2009-12-17 00:00:00	0.28269	0.2750	0.2400	...	-24.20	0.0	0.0	0.0	0.0	1.0	0	0	0.0	41
3	0EF5356002482715299901A	658116	2012-10-22 11:02:35.010000000	NaN	36	Current	NaN	0.12528	0.0974	0.0874	...	-108.01	0.0	0.0	0.0	0.0	1.0	0	0	0.0	158
4	0F023589499656230C5E3E2	909464	2013-09-14 18:38:39.097000000	NaN	36	Current	NaN	0.24614	0.2085	0.1985	...	-60.27	0.0	0.0	0.0	0.0	1.0	0	0	0.0	20

5 rows × 81 columns

# Few variables are used for the project to simplify analysis:  
cols = ['ListingKey',
        'Term' , 
        'LoanStatus', 
        'EstimatedEffectiveYield' ,  
        'BorrowerAPR' , 
        'BorrowerRate' , 
        'ProsperRating (numeric)',
        'ProsperRating (Alpha)',
        'ProsperScore',                     
        'ListingCategory (numeric)', 
        'EmploymentStatus',
        'Occupation', 
        'EmploymentStatusDuration', 
        'IsBorrowerHomeowner', 
        'IncomeVerifiable', 
        'StatedMonthlyIncome', 
        'MonthlyLoanPayment',
        'Recommendations', 
        'DebtToIncomeRatio',
        'LoanOriginalAmount' ,
        'PercentFunded',
        'IncomeRange',
        'Investors',
        'BorrowerState']

loan_data = df[cols]
display(loan_data.head())

	ListingKey	Term	LoanStatus	EstimatedEffectiveYield	BorrowerAPR	BorrowerRate	ProsperRating (numeric)	ProsperRating (Alpha)	ProsperScore	ListingCategory (numeric)	...	IncomeVerifiable	StatedMonthlyIncome	MonthlyLoanPayment	Recommendations	DebtToIncomeRatio	LoanOriginalAmount	PercentFunded	IncomeRange	Investors	BorrowerState
0	1021339766868145413AB3B	36	Completed	NaN	0.16516	0.1580	NaN	NaN	NaN	0	...	True	3083.333333	330.43	0	0.17	9425	1.0	$25,000-49,999	258	CO
1	10273602499503308B223C1	36	Current	0.07960	0.12016	0.0920	6.0	A	7.0	2	...	True	6125.000000	318.93	0	0.18	10000	1.0	$50,000-74,999	1	CO
2	0EE9337825851032864889A	36	Completed	NaN	0.28269	0.2750	NaN	NaN	NaN	0	...	True	2083.333333	123.32	0	0.06	3001	1.0	Not displayed	41	GA
3	0EF5356002482715299901A	36	Current	0.08490	0.12528	0.0974	6.0	A	9.0	16	...	True	2875.000000	321.45	0	0.15	10000	1.0	$25,000-49,999	158	GA
4	0F023589499656230C5E3E2	36	Current	0.18316	0.24614	0.2085	3.0	D	4.0	2	...	True	9583.333333	563.97	0	0.26	15000	1.0	$100,000+	20	MN

5 rows × 24 columns

display(loan_data.info())
display(loan_data.isnull().sum())
print('duplicate rows:',loan_data.duplicated().sum())

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 113937 entries, 0 to 113936
Data columns (total 24 columns):
 #   Column                     Non-Null Count   Dtype  
---  ------                     --------------   -----  
 0   ListingKey                 113937 non-null  object 
 1   Term                       113937 non-null  int64  
 2   LoanStatus                 113937 non-null  object 
 3   EstimatedEffectiveYield    84853 non-null   float64
 4   BorrowerAPR                113912 non-null  float64
 5   BorrowerRate               113937 non-null  float64
 6   ProsperRating (numeric)    84853 non-null   float64
 7   ProsperRating (Alpha)      84853 non-null   object 
 8   ProsperScore               84853 non-null   float64
 9   ListingCategory (numeric)  113937 non-null  int64  
 10  EmploymentStatus           111682 non-null  object 
 11  Occupation                 110349 non-null  object 
 12  EmploymentStatusDuration   106312 non-null  float64
 13  IsBorrowerHomeowner        113937 non-null  bool   
 14  IncomeVerifiable           113937 non-null  bool   
 15  StatedMonthlyIncome        113937 non-null  float64
 16  MonthlyLoanPayment         113937 non-null  float64
 17  Recommendations            113937 non-null  int64  
 18  DebtToIncomeRatio          105383 non-null  float64
 19  LoanOriginalAmount         113937 non-null  int64  
 20  PercentFunded              113937 non-null  float64
 21  IncomeRange                113937 non-null  object 
 22  Investors                  113937 non-null  int64  
 23  BorrowerState              108422 non-null  object 
dtypes: bool(2), float64(10), int64(5), object(7)
memory usage: 19.3+ MB

None

ListingKey                       0
Term                             0
LoanStatus                       0
EstimatedEffectiveYield      29084
BorrowerAPR                     25
BorrowerRate                     0
ProsperRating (numeric)      29084
ProsperRating (Alpha)        29084
ProsperScore                 29084
ListingCategory (numeric)        0
EmploymentStatus              2255
Occupation                    3588
EmploymentStatusDuration      7625
IsBorrowerHomeowner              0
IncomeVerifiable                 0
StatedMonthlyIncome              0
MonthlyLoanPayment               0
Recommendations                  0
DebtToIncomeRatio             8554
LoanOriginalAmount               0
PercentFunded                    0
IncomeRange                      0
Investors                        0
BorrowerState                 5515
dtype: int64

duplicate rows: 0

Conclusion:

In the loan_data dataset, there have 113937 rows and 24 columns.All the data types are correct.There have missing values in the EstimatedEffectiveYield,BorrowerAPR , ProsperRating (numeric),ProsperRating (Alpha) , ProsperScore , EmploymentStatus , Occupation ,EmploymentStatusDuration,DebtToIncomeRatio,BorrowerState columns.

As this project is all about visualization, for simplicity purpose null values are deleted.

#convert term into category
loan_data['Term'] = loan_data['Term'].astype('category')
loan_data = loan_data.dropna()
display(loan_data.info())

<class 'pandas.core.frame.DataFrame'>
Int64Index: 76216 entries, 1 to 113936
Data columns (total 24 columns):
 #   Column                     Non-Null Count  Dtype   
---  ------                     --------------  -----   
 0   ListingKey                 76216 non-null  object  
 1   Term                       76216 non-null  category
 2   LoanStatus                 76216 non-null  object  
 3   EstimatedEffectiveYield    76216 non-null  float64 
 4   BorrowerAPR                76216 non-null  float64 
 5   BorrowerRate               76216 non-null  float64 
 6   ProsperRating (numeric)    76216 non-null  float64 
 7   ProsperRating (Alpha)      76216 non-null  object  
 8   ProsperScore               76216 non-null  float64 
 9   ListingCategory (numeric)  76216 non-null  int64   
 10  EmploymentStatus           76216 non-null  object  
 11  Occupation                 76216 non-null  object  
 12  EmploymentStatusDuration   76216 non-null  float64 
 13  IsBorrowerHomeowner        76216 non-null  bool    
 14  IncomeVerifiable           76216 non-null  bool    
 15  StatedMonthlyIncome        76216 non-null  float64 
 16  MonthlyLoanPayment         76216 non-null  float64 
 17  Recommendations            76216 non-null  int64   
 18  DebtToIncomeRatio          76216 non-null  float64 
 19  LoanOriginalAmount         76216 non-null  int64   
 20  PercentFunded              76216 non-null  float64 
 21  IncomeRange                76216 non-null  object  
 22  Investors                  76216 non-null  int64   
 23  BorrowerState              76216 non-null  object  
dtypes: bool(2), category(1), float64(10), int64(4), object(7)
memory usage: 13.0+ MB

<ipython-input-6-b7fb12510a03>:2: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  loan_data['Term'] = loan_data['Term'].astype('category')

None

#column label change
loan_data.rename(columns={'ProsperRating (Alpha)': 'ProsperRating_Alpha', 
                        'ProsperRating (numeric)': 'ProsperRating_Numeric',
                        'ListingCategory (numeric)' : 'ListingCategory_Numeric'
                       }, inplace=True)

#making column names lower letter
loan_data.columns=loan_data.columns.str.lower()
display(loan_data.head(1))

	listingkey	term	loanstatus	estimatedeffectiveyield	borrowerapr	borrowerrate	prosperrating_numeric	prosperrating_alpha	prosperscore	listingcategory_numeric	...	incomeverifiable	statedmonthlyincome	monthlyloanpayment	recommendations	debttoincomeratio	loanoriginalamount	percentfunded	incomerange	investors	borrowerstate
1	10273602499503308B223C1	36	Current	0.0796	0.12016	0.092	6.0	A	7.0	2	...	True	6125.0	318.93	0	0.18	10000	1.0	$50,000-74,999	1	CO

1 rows × 24 columns

display(loan_data['incomerange'].unique())

array(['$50,000-74,999', '$25,000-49,999', '$100,000+', '$75,000-99,999',
       '$1-24,999', 'Not employed'], dtype=object)

#income_group function have been created to categorize the income level.

def income_group(row):
    
    incomerange = row['incomerange']
    
    if incomerange == "$1-24,999":
        return 'low income '
        
    if incomerange == "$25,000-49,999":
        return "low middle income"
        
    if incomerange == "$50,000-74,999":
        return "middle middle income"
        
    if incomerange == "$75,000-999,999":
        return "upper middle income" 
    
    if incomerange == "$100,000+":
        return "high income" 
    
    if incomerange == "Not employed":
        return "job less" 

loan_data['income label'] = loan_data.apply(income_group, axis=1 ).sort_values(ascending=False)
display(loan_data.head())

	listingkey	term	loanstatus	estimatedeffectiveyield	borrowerapr	borrowerrate	prosperrating_numeric	prosperrating_alpha	prosperscore	listingcategory_numeric	...	statedmonthlyincome	monthlyloanpayment	recommendations	debttoincomeratio	loanoriginalamount	percentfunded	incomerange	investors	borrowerstate	income label
1	10273602499503308B223C1	36	Current	0.07960	0.12016	0.0920	6.0	A	7.0	2	...	6125.000000	318.93	0	0.18	10000	1.0	$50,000-74,999	1	CO	middle middle income
3	0EF5356002482715299901A	36	Current	0.08490	0.12528	0.0974	6.0	A	9.0	16	...	2875.000000	321.45	0	0.15	10000	1.0	$25,000-49,999	158	GA	low middle income
4	0F023589499656230C5E3E2	36	Current	0.18316	0.24614	0.2085	3.0	D	4.0	2	...	9583.333333	563.97	0	0.26	15000	1.0	$100,000+	20	MN	high income
5	0F05359734824199381F61D	60	Current	0.11567	0.15425	0.1314	5.0	B	10.0	1	...	8333.333333	342.37	0	0.36	15000	1.0	$100,000+	1	NM	high income
6	0F0A3576754255009D63151	36	Current	0.23820	0.31032	0.2712	2.0	E	2.0	1	...	2083.333333	122.67	0	0.27	3000	1.0	$25,000-49,999	1	KS	low middle income

5 rows × 25 columns

Univariate Exploration

LoanStatus of all Borrowers

loan_status = loan_data.pivot_table(index = 'loanstatus', values = 'listingkey', aggfunc = 'nunique').reset_index()
loan_status.rename(columns = {'listingkey':'count'}, inplace = True)
loan_status.sort_values(by='count',ascending=False,inplace=True)
fig = px.bar(loan_status, x='loanstatus', y='count', 
             title='Loan status of all borrowers')
fig.show() 

Conclusion:

LoanStatus of all Borrowers are with current and completed state.

EmploymentStatus of all Borrowers

employment_status = loan_data.pivot_table(index = 'employmentstatus', values = 'listingkey', aggfunc = 'nunique').reset_index()
employment_status.rename(columns = {'listingkey':'count'}, inplace = True)
employment_status.sort_values(by='count',ascending=False,inplace=True)
fig = px.bar(employment_status, x='employmentstatus', y='count', 
             title='Employment_status status of all borrowers')
fig.show() 

Conclusion:

EmploymentStatus of all Borrowers are with Employed State and most of them are full time worker.

IncomeRange of all Borrowers

income_range = loan_data.pivot_table(index = 'income label', values = 'listingkey', aggfunc = 'nunique').reset_index()
income_range .rename(columns = {'listingkey':'count'}, inplace = True)
income_range.sort_values(by='count',ascending=False,inplace=True)
fig = px.bar(income_range , x= 'income label', y='count', 
             title='Income status of all borrowers')
fig.show() 

Conclusion:

People having middle middle income(50,000-74,999 USD) and low middle income (25,000-49,999 USD) tool more loans.Job less and low income people have less chance to get loans from bank.

Top 10 states of all Borrowers

state_top10 = loan_data.pivot_table(index = 'borrowerstate', values = 'listingkey', aggfunc = 'nunique').reset_index()
state_top10 .rename(columns = {'listingkey':'count'}, inplace = True)
state_top10.sort_values(by='count',ascending=False,inplace=True)
top10_state = state_top10.head(10)
fig = px.bar(top10_state , x= 'borrowerstate', y='count', 
             title='Top 10 states of all Borrowers')
fig.show() 

Conclusion:

Top 5 states of all Borrowers are from CA,NY,TX,FL and IL

Top 10 - Occupation of all Borrowers

ocuupation_top10 = loan_data.pivot_table(index = 'occupation', values = 'listingkey', aggfunc = 'nunique').reset_index()
ocuupation_top10.rename(columns = {'listingkey':'count'}, inplace = True)
ocuupation_top10.sort_values(by='count',ascending=False,inplace=True)
top10_occupation = ocuupation_top10.head(10)
fig = px.bar(top10_occupation, x= 'occupation', y='count', 
             title='Top 10 occupation of all Borrowers')
fig.show() 

Conclusion:

Most of the borrowers occupation are not defined.May be self employed like property owner.But majority are with an occupation of Professional and Executive.

ProsperScore of all Borrowers

prosperscore = loan_data.pivot_table(index = 'prosperscore', values = 'listingkey', aggfunc = 'nunique').reset_index()
prosperscore .rename(columns = {'listingkey':'count'}, inplace = True)
prosperscore .sort_values(by='count',ascending=False,inplace=True)
fig = px.bar(prosperscore , x= 'prosperscore', y='count', 
             title='ProsperScore of all Borrowers')
fig.show() 

Conclusion:

Majority of the borrowers are with a rating or score from 4 to 8.They have higher chance to approve loan.

Numeric attributes analyze :

The variables that are numeric are 'term', 'estimatedeffectiveyield', 'borrowerapr', 'borrowerrate', 'prosperrating_numeric', 'prosperscore', 'listingcategory_numeric', 'employmentstatusduration', 'statedmonthlyincome', 'monthlyloanpayment', 'recommendations', 'debtToIncomeratio', 'loanoriginalamount', 'percentfunded', 'investors'

cols_to_plot = loan_data[['term', 'estimatedeffectiveyield', 'borrowerapr', 'borrowerrate', 'prosperrating_numeric', 
                'prosperscore', 'listingcategory_numeric', 'employmentstatusduration', 'statedmonthlyincome', 
                'monthlyloanpayment', 'recommendations', 'debttoincomeratio', 'loanoriginalamount', 
                'percentfunded', 'investors']]

display(cols_to_plot.head())

	term	estimatedeffectiveyield	borrowerapr	borrowerrate	prosperrating_numeric	prosperscore	listingcategory_numeric	employmentstatusduration	statedmonthlyincome	monthlyloanpayment	recommendations	debttoincomeratio	loanoriginalamount	percentfunded	investors
1	36	0.07960	0.12016	0.0920	6.0	7.0	2	44.0	6125.000000	318.93	0	0.18	10000	1.0	1
3	36	0.08490	0.12528	0.0974	6.0	9.0	16	113.0	2875.000000	321.45	0	0.15	10000	1.0	158
4	36	0.18316	0.24614	0.2085	3.0	4.0	2	44.0	9583.333333	563.97	0	0.26	15000	1.0	20
5	60	0.11567	0.15425	0.1314	5.0	10.0	1	82.0	8333.333333	342.37	0	0.36	15000	1.0	1
6	36	0.23820	0.31032	0.2712	2.0	2.0	1	172.0	2083.333333	122.67	0	0.27	3000	1.0	1

sns.boxplot(x=loan_data['borrowerrate'])

<AxesSubplot:xlabel='borrowerrate'>

loan_data.hist('borrowerrate', bins = 10);
plt.xlabel('borrower interest rate');
plt.ylabel('Count');
plt.title("borrower interest rate of all Borrowers");

sns.boxplot(x=loan_data['statedmonthlyincome'])

<AxesSubplot:xlabel='statedmonthlyincome'>

loan_data_clean = loan_data.copy()

#Determine the upper and lower limits of outliers and remove the outliers(statedmonthlyincome) using IQR rule

uw_1 =(loan_data_clean['statedmonthlyincome'].quantile(0.75))+(1.5*((loan_data_clean['statedmonthlyincome'].quantile(0.75))-(loan_data_clean['statedmonthlyincome'].quantile(0.25))))
lw_1=(loan_data_clean['statedmonthlyincome'].quantile(0.25))-(1.5*((loan_data_clean['statedmonthlyincome'].quantile(0.75))-(loan_data_clean['statedmonthlyincome'].quantile(0.25))))
print(uw_1)
print(lw_1)
loan_data_clean = loan_data_clean.query(' @lw_1 <= statedmonthlyincome <= @uw_1 ')
display(loan_data_clean.head(2))

12750.0000005
-1916.6666675000001

	listingkey	term	loanstatus	estimatedeffectiveyield	borrowerapr	borrowerrate	prosperrating_numeric	prosperrating_alpha	prosperscore	listingcategory_numeric	...	statedmonthlyincome	monthlyloanpayment	recommendations	debttoincomeratio	loanoriginalamount	percentfunded	incomerange	investors	borrowerstate	income label
1	10273602499503308B223C1	36	Current	0.0796	0.12016	0.0920	6.0	A	7.0	2	...	6125.0	318.93	0	0.18	10000	1.0	$50,000-74,999	1	CO	middle middle income
3	0EF5356002482715299901A	36	Current	0.0849	0.12528	0.0974	6.0	A	9.0	16	...	2875.0	321.45	0	0.15	10000	1.0	$25,000-49,999	158	GA	low middle income

2 rows × 25 columns

sns.boxplot(x=loan_data_clean['statedmonthlyincome'])

<AxesSubplot:xlabel='statedmonthlyincome'>

loan_data_clean.hist('statedmonthlyincome', bins = 20);
plt.xlabel('StatedMonthlyIncome');
plt.ylabel('Count');
plt.title("StatedMonthlyIncome of all Borrowers");

sns.boxplot(x=loan_data['loanoriginalamount'])

<AxesSubplot:xlabel='loanoriginalamount'>

#Determine the upper and lower limits of outliers and remove the outliers(loanoriginalamount) using IQR rule

uw_2 =(loan_data_clean['loanoriginalamount'].quantile(0.75))+(1.5*((loan_data_clean['loanoriginalamount'].quantile(0.75))-(loan_data_clean['loanoriginalamount'].quantile(0.25))))
lw_2=(loan_data_clean['loanoriginalamount'].quantile(0.25))-(1.5*((loan_data_clean['loanoriginalamount'].quantile(0.75))-(loan_data_clean['loanoriginalamount'].quantile(0.25))))
print(uw_2)
print(lw_2)
loan_data_clean = loan_data_clean.query(' @lw_2 <= loanoriginalamount <= @uw_2 ')
display(loan_data_clean.head(2))

26500.0
-9500.0

	listingkey	term	loanstatus	estimatedeffectiveyield	borrowerapr	borrowerrate	prosperrating_numeric	prosperrating_alpha	prosperscore	listingcategory_numeric	...	statedmonthlyincome	monthlyloanpayment	recommendations	debttoincomeratio	loanoriginalamount	percentfunded	incomerange	investors	borrowerstate	income label
1	10273602499503308B223C1	36	Current	0.0796	0.12016	0.0920	6.0	A	7.0	2	...	6125.0	318.93	0	0.18	10000	1.0	$50,000-74,999	1	CO	middle middle income
3	0EF5356002482715299901A	36	Current	0.0849	0.12528	0.0974	6.0	A	9.0	16	...	2875.0	321.45	0	0.15	10000	1.0	$25,000-49,999	158	GA	low middle income

2 rows × 25 columns

sns.boxplot(x=loan_data_clean['loanoriginalamount'])

<AxesSubplot:xlabel='loanoriginalamount'>

loan_data_clean.hist('loanoriginalamount', bins = 20);
plt.xlabel('loan originalamount');
plt.ylabel('Count');
plt.title("loan original amount of all Borrowers");

sns.boxplot(x=loan_data['employmentstatusduration'])

<AxesSubplot:xlabel='employmentstatusduration'>

#Determine the upper and lower limits of outliers and remove the outliers(employmentstatusduration) using IQR rule

uw_3 =(loan_data_clean['employmentstatusduration'].quantile(0.75))+(1.5*((loan_data_clean['employmentstatusduration'].quantile(0.75))-(loan_data_clean['employmentstatusduration'].quantile(0.25))))
lw_3=(loan_data_clean['employmentstatusduration'].quantile(0.25))-(1.5*((loan_data_clean['employmentstatusduration'].quantile(0.75))-(loan_data_clean['employmentstatusduration'].quantile(0.25))))
print(uw_3)
print(lw_3)
loan_data_clean = loan_data_clean.query(' @lw_3 <=employmentstatusduration <= @uw_3')
display(loan_data_clean.head(2))

324.5
-143.5

	listingkey	term	loanstatus	estimatedeffectiveyield	borrowerapr	borrowerrate	prosperrating_numeric	prosperrating_alpha	prosperscore	listingcategory_numeric	...	statedmonthlyincome	monthlyloanpayment	recommendations	debttoincomeratio	loanoriginalamount	percentfunded	incomerange	investors	borrowerstate	income label
1	10273602499503308B223C1	36	Current	0.0796	0.12016	0.0920	6.0	A	7.0	2	...	6125.0	318.93	0	0.18	10000	1.0	$50,000-74,999	1	CO	middle middle income
3	0EF5356002482715299901A	36	Current	0.0849	0.12528	0.0974	6.0	A	9.0	16	...	2875.0	321.45	0	0.15	10000	1.0	$25,000-49,999	158	GA	low middle income

2 rows × 25 columns

sns.boxplot(x=loan_data_clean['employmentstatusduration'])

<AxesSubplot:xlabel='employmentstatusduration'>

loan_data_clean.hist('employmentstatusduration', bins = 30);
plt.xlabel('employment status duration');
plt.ylabel('Count');
plt.title("Employment status duration");

display(loan_data[['statedmonthlyincome','loanoriginalamount','employmentstatusduration']].describe())

	statedmonthlyincome	loanoriginalamount	employmentstatusduration
count	76216.000000	76216.000000	76216.000000
mean	6002.833586	9294.999410	104.578028
std	5110.743506	6413.404265	95.836443
min	0.250000	1000.000000	0.000000
25%	3583.333333	4000.000000	32.000000
50%	5041.666667	8000.000000	76.000000
75%	7250.000000	14603.250000	150.000000
max	483333.333333	35000.000000	755.000000

display(loan_data[['borrowerrate']].describe())

	borrowerrate
count	76216.000000
mean	0.193621
std	0.074088
min	0.040000
25%	0.134900
50%	0.184500
75%	0.254900
max	0.360000

Insights with Numerical variables analysis:

BorrowerRate : The average interest rate is 0.19.The maximum interest rate is 0.36 and minimum rate is 0.04.But most of the borrowers inetrest rate is 0.16.

StatedMonthlyIncome : The average monthly income is approx. 6002 USD.The maximum income is 483333 USD and minimum rate is 0.25.But most of the borrowers monthky incomw 4500 USD.

LoanOriginalAmount: The average amount is 9294 USD.The maximum loan is 35000 USD and minimum is 1000.But most of the borrowers loan amaount is approx. 4500 USD.

Employment status duration: The average amount is 104.5 months.The maximum is 755 months and the minimum is 0.But most of the borrowers took loan whose have employement status 0-50 months.

Bivariate Exploration:

Relationship between two numeric variables

loan_data_clean.dtypes

listingkey                    object
term                        category
loanstatus                    object
estimatedeffectiveyield      float64
borrowerapr                  float64
borrowerrate                 float64
prosperrating_numeric        float64
prosperrating_alpha           object
prosperscore                 float64
listingcategory_numeric        int64
employmentstatus              object
occupation                    object
employmentstatusduration     float64
isborrowerhomeowner             bool
incomeverifiable                bool
statedmonthlyincome          float64
monthlyloanpayment           float64
recommendations                int64
debttoincomeratio            float64
loanoriginalamount             int64
percentfunded                float64
incomerange                   object
investors                      int64
borrowerstate                 object
income label                  object
dtype: object

numeric_vars = ['estimatedeffectiveyield', 'borrowerapr', 'borrowerrate',
                'prosperrating_numeric', 'prosperscore', 
                'listingcategory_numeric','employmentstatusduration', 
                'statedmonthlyincome', 'monthlyloanpayment',
                'recommendations', 'debttoincomeratio', 'loanoriginalamount','percentfunded', 'investors']

#numeric_vars_1 = ['BorrowerRate',
                # 'ProsperScore',                 
                #'StatedMonthlyIncome', 'MonthlyLoanPayment',
                #'LoanOriginalAmount', 'Investors' ,'LoanOriginalAmount_log',
                 #'Investors_log']

categoric_vars = ['loanStatus', 'prosperrating_alpha', 'employmentstatus', 'occupation',
       'incomerange', 'borrowerstate', 'term', 'isborrowerhomeowner']

# correlation plot
plt.figure(figsize = [12, 8])
sns.heatmap(loan_data_clean[numeric_vars].corr(), annot = True, fmt = '.3f',
           cmap = 'vlag_r', center = 0)
plt.show()

*Conclusion:

1. Loan original amount and monthly loan payment is highly correlated.
2. Borrower annual percentage rate and prosper score is negatively correlated.

Multivariate Exploration

Applicants rating with monthly income and Employment Status

# To identify how the applicants rating is affected by Employment status and Monthly income
plt.figure(figsize = [15, 7])
ax = sns.pointplot(data = loan_data_clean, x = 'employmentstatus', y = 'statedmonthlyincome', hue = 'prosperrating_numeric',
                  dodge = 0.3, linestyles = "")
plt.title('Applicants - Rating across Monthly Income and EmploymentStatus');

Conclusion:

1. For Applicants(employed and fulltime) with prosper ratings from 7 to 4 have the higher loan amount with increased salary.
2. For Applicants(parttime employee) with prosper ratings from 7 to 4 have the lower loan amount with low level salary.

Did homeowner status has impacted the ProsperRating and Borrower Interest Rate?

plt.figure(figsize = [15, 7])
ax = sns.pointplot(data = loan_data_clean, x = 'prosperrating_numeric', y = 'borrowerrate', hue = 'isborrowerhomeowner',
                  dodge = 0.3, linestyles = "");
plt.title('Applicants - Home ownner status across ProsperRating and BorrowerRate ');

Conclusion:

We observe that without homeowner tend to have a higher interest rate, and thus lower rating.However homeowner tends to have lower interest rate and higher rating. So we can safely say that homeowner is safest bet when giving a loan.

Overall findings:

To summarize this report, I believe that the loan approval status is heavily influenced by the applicant's details on income range, house owner status, and job status.